Fighting selection bias in statistical learning: application to visual recognition from biased image databases - Télécom Paris Accéder directement au contenu
Article Dans Une Revue Journal of Nonparametric Statistics Année : 2023

Fighting selection bias in statistical learning: application to visual recognition from biased image databases

Résumé

In practice, and especially when training deep neural networks, visual recognition rules are often learned based on various sources of information. On the other hand, the recent deployment of facial recognition systems with uneven performances on different population segments has highlighted the representativeness issues induced by a naive aggregation of the datasets. In this paper, we show how biasing models can remedy these problems. Based on the (approximate) knowledge of the biasing mechanisms at work, our approach consists in reweighting the observations, so as to form a nearly debiased estimator of the target distribution. One key condition is that the supports of the biased distributions must partly overlap, and cover the support of the target distribution. In order to meet this requirement in practice, we propose to use a low dimensional image representation, shared across the image databases. Finally, we provide numerical experiments highlighting the relevance of our approach

Dates et versions

hal-04267977 , version 1 (02-11-2023)

Identifiants

Citer

Stéphan Clémençon, Pierre Laforgue, Robin Vogel. Fighting selection bias in statistical learning: application to visual recognition from biased image databases. Journal of Nonparametric Statistics, 2023, pp.1-24. ⟨10.1080/10485252.2023.2259011⟩. ⟨hal-04267977⟩
18 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More